Static and dynamic non-localized efficiency radio frequency shimming for parallel transmission in magnetic resonance imaging

Abstract

A non-localized efficiency shimming technique is used to generate radio frequency (RF) shimming values for imaging with a multi-channel transmit RF coil that minimizes subject-specific imperfections in the transmit magnetic field (B1+) and reduces or eliminates signal dropout in the acquired images, while keeping the coil working in an optimal mode with a high transmit efficiency. The non-localized efficiency shimming can be used for both small and large fields-of-view where a specific ROI does not need to be specified. The static non-localized efficiency shim is advantageous for turbo spin echo (TSE) imaging of smaller anatomical targets, whereas the dynamic non-localized efficiency shim is advantageous for larger fields-of-view, such as in human torsos.

The invention claimed is: 
     
         1 . A method for generating radio frequency (RF) shimming values for use with a magnetic resonance imaging system, the method comprising:
 (a) accessing RF transmit sensitivity profile data with a computer system;   (b) generating RF shimming values with the computer system by:
 inputting the RF transmit sensitivity profile data to a non-localized efficiency shimming cost function that penalizes under-flipping based on a minimum flip angle tuning parameter; and 
 optimizing the non-localized efficiency shimming cost function, generating an output as the RF shimming values, wherein the RF shimming values minimize destructive B1+ interferences within an entire imaging field-of-view; and 
   (c) storing the RF shimming values for use with an MRI system.   
     
     
         2 . The method of  claim 1 , wherein the non-localized efficiency shimming cost function further optimizes for maximum flip angle across modes. 
     
     
         3 . The method of  claim 1 , wherein optimizing the non-localized efficiency shimming cost function includes setting at least one constraint on the non-localized efficiency shimming cost function while the non-localized efficiency shimming cost function is being optimized. 
     
     
         4 . The method of  claim 1 , wherein the non-localized efficiency shimming cost function penalizes under-flipping without explicitly constraining over-flipping. 
     
     
         5 . The method of  claim 1 , further comprising:
 accessing the RF shimming values with an MRI system;   controlling the MRI system to acquire k-space data using a pulse sequence that implements the RF shimming values; and   reconstructing an image from the k-space data.   
     
     
         6 . The method of  claim 5 , wherein the pulse sequence includes a gradient-recalled echo (GRE) acquisition. 
     
     
         7 . The method of  claim 1 , wherein the RF transmit sensitivity profile data comprises RF transmit sensitivity profiles for a plurality of subjects and the non-localized efficiency shimming cost function incorporates the RF transmit sensitivity profiles for a plurality of subjects to generate universal modes. 
     
     
         8 . The method of  claim 2 , wherein the non-localized efficiency shimming cost function further optimizes for maximum flip angle across modes based in part on a maximum flip angle tuning parameter. 
     
     
         9 . The method of  claim 3 , wherein the at least one constraint comprises a constraint for a desired efficiency in a local region-of-interest using a Rayleigh quotient. 
     
     
         10 . The method of  claim 3 , wherein the at least one constraint comprises a constraint for local specific absorption rate (SAR) based on virtual observation points. 
     
     
         11 . The method of  claim 5 , wherein the pulse sequence includes a spin echo acquisition. 
     
     
         12 . The method of  claim 7 , wherein the non-localized efficiency shimming cost function is optimized across the plurality of subjects. 
     
     
         13 . The method of  claim 9 , wherein the Rayleigh quotient is used as a starting point for performing a phase-only RF shim. 
     
     
         14 . The method of  claim 11 , wherein the spin echo acquisition is a turbo spin echo (TSE) acquisition. 
     
     
         15 . The method of  claim 11 , wherein the spin echo acquisition is a fast spin echo (FSE) acquisition. 
     
     
         16 . The method of  claim 12 , wherein optimizing the non-localized efficiency shimming cost function across the plurality of subjects includes constructing a plurality of non-localized efficiency shimming cost functions comprises a different non-localized efficiency shimming cost function for each of the plurality of subjects, and minimizing a Euclidian norm of the plurality of non-localized efficiency shimming cost functions.